224 
Fishery Bulletin 95(2), 1997 
1965 1970 1975 1980 1985 1990 1995 
Figure 3 
Catch per hour (kg) of (A) the dominant species of rockfish (Pacific ocean 
perch and rougheye rockfish), and (B) shortspine thornyheads at the north- 
ern Washington index sites, 1968-92. Error bars indicate approximate 95% 
confidence intervals (2SE). 
differences in annual growth in all ages younger than 
15. The magnitude of interannual differences in size 
at age 15 for a given depth and sex were relatively 
minor (Fig. 8), particularly when differences in age 
determination technique are considered. 
The length-maturity curve for 1992 (Fig. 9) indi- 
cates a significant shift toward maturation at a 
smaller length than was the case in 1968-72 (Table 
5; Fig. 9). Both a Z-test (Table 5) and logistic regres- 
sion analysis (SPSS) showed that the year effect was 
highly significant (PcO.OOl) statistically. Age at 50% 
maturity, as predicted from the Bertalanffy growth 
model presented in Gunderson (1977, Table 3) de- 
creased from 10.1 to 8.1 years over this same period. 
Discussion 
Although members of the genus Sebastes are vivipa- 
rous and retain their larvae for varying periods prior 
to releasing them, the larvae can still be transported 
considerable distances during their 3-6 month pe- 
lagic phase, resulting in broad spatial synchrony in 
recruitment trends (Ralston and Howard, 1995). In 
